La rutina de silvia answers

Article Summary

https://doi.org/10.1080/09581596.2018.1444267

COMMENTARY

The opportunity cost of pharmaceutical price increases: improving health by investing in education

Jonathan E. Fieldinga, Frederick J. Zimmermana and Kristin Calsadab

aCenter for Health advancement, Department of Health Policy & Management, Fielding School of Public Health, uCla, los angeles, Ca, uSa; bDepartment of Health Policy & Management, Fielding School of Public Health, uCla, los angeles, Ca, uSa

ABSTRACT A federal law prohibits the US Government from negotiating pharmaceutical prices. This law comes with an opportunity cost: resources spent on unnecessarily highly priced drugs cannot be spent on other social goals. To calculate the opportunity cost of this spending, this analysis first identified a proxy for unnecessarily high pharmaceutical spending. We then estimated the value of the outcomes which this money would produce if invested in an alternative, high-value use. We estimated the excess price increases in a set of 80 commonly prescribed drugs paid for by the Centers for Medicare and Medicaid Services from 2010 to 2014. The value of price increases among these drugs above the rate of medical inflation was $11.5 billion dollars. This money has alternative uses, including some that promote health and other social goals. This is the opportunity cost of unnecessarily high pharmaceutical spending. Investment in high-school dropout prevention programs was chosen as a measure of alternative uses for this spending because of the importance of education as a social determinant of health and because medical spending has been shown to specifically crowd out education spending. Invested in programs to increase high-school graduation rates, this money could create an additional 200,000 high-school graduates, which in turn would generate an estimated $32 billion in returns (net present value) to government and health improvements of up to 1 million quality-adjusted life years (QALYs) per year of redirected expenditures.

Introduction

In 2014, medical care-related expenditures in the United States accounted for 17.8% of gross domestic product (GDP), totaling over $3 trillion (Centers for Medicare & Medicaid Services, 2015b). Yet despite outspending all other countries on a per capita basis, the United States ranks in the bottom third of non-poor countries in life expectancy at birth and 27th out of 34 in life expectancy at age 60, ahead only of far less wealthy countries like Mexico, Turkey, and Hungary (Scobie, 2015).

A report from the Institute of Medicine (Young & Olsen, 2010) documented $425 billion in excessive costs-per-service delivered in 2009, including $130 billion in inefficiently delivered services, $190 billion in excessive administrative costs, and $105 billion in prices that are too high. Although the report breaks out inefficiency and administrative costs from prices, it should be remembered that all of these factors increase the average cost-per-service without contributing to medical quality or therapeutic benefit.

ARTICLE HISTORY received 8 May 2017 accepted 27 January 2018

KEYWORDS Opportunity cost; social determinants of health; health equity; drug prices

© 2018 informa uK limited, trading as taylor & Francis Group

CONTACT Frederick J. Zimmerman fredzimmerman@ucla.edu

CritiCal PubliC HealtH 2019, Vol. 29, No. 3, 353–362

mailto:fredzimmerman@ucla.edu
http://www.tandfonline.com
http://crossmark.crossref.org/dialog/?doi=10.1080/09581596.2018.1444267&domain=pdf
As a concrete example, one analysis finds that switching to an equally safe and effective drug from one often currently prescribed to prevent certain kinds of blindness could save Medicaid $18 billion over 10 years (Hutton, Newman-Casey, Tavag, Zacks, & Stein, 2014). Continuing to allow the more expensive drug to be prescribed is an administrative inefficiency that works as a de facto price increase.

Although an aging population with a higher disease prevalence and burden contributes to the increase in health expenditures (Thorpe, 2005), prices per unit of product or service far exceed general inflation (The Office of the Actuary in the Centers for Medicare & Medicaid Services, 2017). No doubt some of these price increases have accompanied therapeutic advances, but many have not (Young & Olsen, 2010).

The rapid rate of increase in prescription drug spending has outpaced the price growth of all other medical care products and services. For example, from 2005 to 2014 drug prices increased by 5.4% per annum above the inflation rate for these products and services, after accounting for discounts (Aitken, Berndt, Cutler, Kleinrock, & Maini, 2016).

Economists define the concept of opportunity cost of spending on a given activity as the value that spending would produce if dedicated instead to the next most valuable activity. In the case of medical spending, the question is to what other use such spending might be put, and what is the value – to health and other social goals – of spending on those alternative uses.

An emerging literature (Fossett & Burke, 2004; Tran, Zimmerman, & Fielding, 2017) shows how excess spending from the public purse on medical care displaces public investments in public health, educa- tion, and other social services. All of these alternative uses for public spending have the potential to substantially improve health outcomes. For example, low-income housing vouchers (Gubits et al., 2015), paid family leave (Ruhm, 2000; Tanaka, 2005), universal preschool (Conti & Heckman, 2013), and clean energy (Haines, Kovats, Campbell-Lendrum, & Corvalan, 2006) have the potential to improve health and provide other desired public benefits.

This study calculates excessive spending due to price increases on 80 drugs covered by Medicare as one estimate (among many) of the spending on pharmaceuticals that has little or no therapeutic value. It then discusses how that money could provide better health outcomes if invested in educa- tional programs that increase high-school graduation rates. High-school completion interventions were chosen as an example because strong evidence in the literature suggests that educational attainment improves health status and life expectancy (Conti, Heckman, & Urzua, 2010) and because education spending specifically has been reduced as public spending on medical care has increased (Fossett & Burke, 2004). Using existing cost-effectiveness analyses of several high-school completion programs, the potential health and financial benefits of investing in these programs are calculated.

Increases in Medicare drug prices

Medicare, the second-largest purchaser of prescription drugs in the US, has made publicly available information on the price-per-unit of many of the prescription drugs that it covers. In 2015, the Centers for Medicare and Medicaid Services (CMS) released a Drug Spending Dashboard with data on 40 drugs covered by Medicare Part B and 40 drugs covered by Medicare Part D between 2010 and 2014 (Centers for Medicare & Medicaid Services, 2015a). CMS included these 80 drugs in the dashboard because either (a) they contributed to high total spending (30 drugs), (b) they were not included under (a) but had high annual per-user spending (30 drugs), or (c) they were not included under (a) or (b) but had high price-per-unit increases (20 drugs). In 2014 these 80 drugs collectively accounted for 39% of total drug spending under parts B and D of Medicare. The dashboard displays spending, utilization, price per unit, and drug product descriptions. The spending data for Part D drugs do not reflect manufacturer rebates. Although the rebate amounts are difficult to determine, one recent analysis finds that increases in net prices were 16% slower than increases in invoice prices (Aitken et al., 2016) (e.g. increases of 4.2% instead of 5%).

The 20 drugs included because of high price-per-unit increases accounted for only 3.3% of spending among included drugs (1.3% of Parts B and D spending). For that reason, those drugs are not a major

354 J. E. FIELDING ET AL.

driver of the estimates of price increases in the sample of drugs included. Because of the selection criteria, this list of drugs is not representative of all drugs paid for by Medicare. For that reason, the analysis here is conducted only for this set of drugs, and no attempt is made to make inferences for all drugs. Because of the known difficulties in obtaining information on actual prices paid for drugs, the data-set represents valuable information on the included drugs, which accounts for a significant portion of total spending. As a subset, however, it necessarily underestimates total spending, including spending of zero marginal value.

Among the drugs listed in the dashboard, 90% increased in average price-per-unit between 2010 and 2014, and the average increase across all drugs significantly exceeded both the average medical inflation rate (3.2%) and the average annual rate of general inflation during this period (2%) (Bureau of Labor Statistics, 2016). Price increases in line with inflation are natural, but increases above infla- tion raise suspicion because there is no obvious economic rationale for them. While medical inflation exceeds other inflation because the costs of medical personnel are rising in real terms, it is not clear why the prices of existing drugs – which are basic manufactured items – should be increasing at all above regular inflation.

Of course, real (inflation-adjusted) prices of any good can go up in response to price shocks to inputs. Yet because there have been no major changes in the input prices to pharmaceuticals during this period, the price increases of existing drugs were unlikely to represent market responses. Instead, these price increases likely represent rent-seeking – an economic phenomenon in which companies use patent protection, lobbying, and marketing to secure financial returns above what the free market would provide. The process of increasing the price of existing drugs that are on the market has been called ‘market spiral pricing’ (Light & Kantarjian, 2013) and some researchers assert that it is one example of rent-seeking by pharmaceutical companies (Kantarjian & Zwelling, 2013). The price increases of existing drugs over that five-year period are accordingly one proxy for what a recent Institute of Medicine report labeled as ‘prices that are too high’ (Young & Olsen, 2010).

To quantify this economically unnecessary cost to Medicare, we simulated total spending on drugs listed in the dashboard each year according to two scenarios: (1) average price-per-unit increases at the medical inflation rate and (2) average price-per-unit increases at the general inflation rate. Each simulation adjusted for actual changes in utilization throughout the time period. Then, we calculated the difference between Medicare’s actual spending and the simulated spending for each scenario. Table 1 illustrates the differences in drug spending.

The difference in total spending between actual Medicare spending and simulated projections grew annually in both scenarios except between 2013 and 2014.1

If drug prices had changed at the rate of general inflation between 2010 and 2014, spending of approximately $15.6 billion per year by the 5th year would have been averted. If drug prices had increased at the same rate as medical inflation, the excess spending would still exceed $11.5 billion annually. These excessive expenditures that provide no direct health benefits carry an important oppor- tunity cost.

These estimates are not the only or necessarily the best estimates of zero-marginal-value spending on pharmaceuticals, but they are on the low side of other estimates in the literature. For example, if Medicare were able to negotiate lower drug prices, researchers suggest that it could save between

Table 1. actual and simulated annual medicare drug spending (billions).

2010 2011 2012 2013 2014 Total actual annual spending $23.97 $28.3 $33.9 $42.2 $55.1 $159.5 Spending if prices increased at medical inflation rate – $27.3 $31.5 $36.9 $52.4 $148.0 incremental difference between actual spending and increase due to

medical inflation – $1.0 $2.4 $5.4 $2.7 $11.5

Spending if prices increased with general inflation – $26.8 $30.8 $35.8 $50.4 $143.8 incremental difference between actual spending and increase due to

general inflation – $1.5 $3.1 $6.4 $4.6 $15.6

CRITICAL PUBLIC HEALTH 355

$15.2 billion (Shih, Schwartz, & Coukell, 2016) and nearly $100 billion annually (Baker, 2006). Another analysis found that US consumers and taxpayers pay a premium above the prices paid in other devel- oped countries, and that the value of this premium for the 20 top-selling drugs is $116 billion a year, compared to only $76 billion a year spent (by the 15 companies that produce these drugs) on their global research-and-development budgets (Yu, Helms, & Bach, 2017). Even if one assumes that the US should shoulder the entire burden of R&D, and that this burden should come only out of prices on the top 20 drugs, this implies spending of $40 billion ($116–$76 billion) that provides no current or future health benefit. It has elsewhere been suggested that when one considers all prescription drugs rather than only the top 20, and takes into account more accurate estimates of true R&D costs, this $40 billion is a considerable underestimate (Light, 2017).

The rest of our analysis uses the $11.5 billion estimate. This estimate is not necessarily the best; it is used because it is the most conservative credible estimate.

Education as an alternative investment

The billions of excess dollars spent due to pharmaceutical price increases could provide better health returns if this money were instead invested in a social determinant of health, such as education. Education is chosen because other research has shown that medical spending crowds out spending on education, at least at the state level (Fossett & Burke, 2004), and because education (1) increases health knowledge and healthy behaviors, (2) enhances employment opportunities and fosters higher income, and (3) enhances social and psychosocial factors, such as social support, executive function, and perceptions of self-control, that affect health (Brunello, Fort, Schneeweis, & Winter-Ebmer, 2016; Ross & Wu, 1995). Although formal educational attainment can be considered a social determinant of health, it is provided as a social service through a process that is defined at the population level (Tanenbaum, 2017). In a sense, society makes a decision about how many of its people will have various levels of education. It is in this sense – educational outcomes as a collective choice about resource allocation – that education is understood here, not as a manifestation of individual achievement.

The opportunity cost of excessive spending can accordingly be estimated by calculating the financial benefits to government and the health benefits to society from an investment in educational programs that increases high-school completion rates.

A meta-analysis conducted by Wilson, Tanner-Smith, Lipsey, Steinka-Fry, and Morrison (2011) and a systematic review by the US Community Preventive Services Task Force identified over 300 programs designed to increase high-school completion rates (The Guide to Community Preventive Services, 2013). To identify the relationship between the costs required to implement interventions and their potential health and financial benefits, we restricted our analysis to programs that were implemented in the United States and reported data on cost of implementation. Interventions that specifically targeted General Education Diploma (GED) attainment were not included because the correlation between GED attainment and health status is not definitively studied in the literature. This process of exclusion resulted in a subset of 17 interventions with an average odds ratio of 1.81 for high-school graduation.

Among the 17 interventions, there were wide variations in methods, effectiveness, cost, location, and population size. For purposes of this economic analysis, we focused on the cost-effectiveness (measured by number of additional graduates per $100,000) of each intervention, conducted from the government’s perspective. On average, the 17 interventions produced 1.76 additional high-school graduates per $100,000 spent, for a cost of $56,300 per additional graduate.

Of these interventions, three are very briefly described in Table 2. Career Academies are small, with- in-school learning communities of 150–200 students that focus on both academic success and technical education, often with an applied component. In several rigorous evaluations, they were found to lower dropout rates for at-risk students (Kemple, 2008; Maxwell & Rubin, 2000). Talent Search is a program that increased high-school graduation rates as a means to increase college matriculation for first-gen- eration college students of low-income families (Levin et al., 2012). Small Schools of Choice are public high-schools located in historically disadvantaged communities (Bloom & Unterman, 2012). They are

J. E. FIELDING ET AL.356

smaller than typical high-schools, and are organized around principles of academic rigor, personalized attention, and job relevance.

Financial returns to government of educational investments

Table 3 illustrates the potential returns in the number of new graduates resulting from each intervention based on the amount of money invested. Current estimates of effectiveness of these three dropout prevention programs in the US suggest a one-time $11.5 billion investment could lead to as many as 500,000 additional high-school graduates (Wilson et al., 2011) (and by extension, an annual investment of $11.5 billion would lead to as many as 500,000 additional graduates every year).

The economic benefits of high-school graduation include increased lifetime income and related increased tax revenue. Careful economic estimates suggest that if a large pool of representative high- school dropouts were instead to graduate from high-school they would earn an average additional $289,820 (2004 dollars) over their lifetime compared to dropouts (Belfield & Levin, 2007a). These increases in income directly translate into increased tax revenue for federal, state, and local governments. Over a lifetime, in discounted present-value terms, this average high-school graduate will contribute an additional $101,180 (2004 dollars; $128,893 in 2016 dollars) in income and sales taxes compared to a dropout (Belfield & Levin, 2007a, p. 53). Table 4 illustrates the estimated lifetime benefits of increased tax revenue expected from a cohort of additional graduates expected to result from a one-time $11.5 billion investment in each of the interventions. Estimates in this table and in the rest of the text are expressed in 2016 dollars, and represent total government costs and cost savings at the local, state, and federal levels.

In addition, high-school graduation is associated with decreased government spending on wel- fare, crime, Medicare, and Medicaid. Due to lower incomes, dropouts are more likely to receive public

Table 2. Costs and effectiveness of three high-school completion interventions (2016 dollars).

Note: Columns (a), (C) and (D) are taken directly from the literature as cited. Columns (b) and (e) are calculated by the authors. Sources: 1belfield and levin (2007b) 2levin et al. (2012) 3bloom and unterman (2012).

Program Description

(A) (B) (C) (D) (E)

Cost per student

Cost per additional

grad

Baseline HS completion

(%)

Percentage increase in HS

completion (%)

Number of new grads

per $100,000 average 17 Programs $7210 $56,300 61.4 12.8 1.77 Career acade-

mies1 School w/in school to

promote employ- ment readiness

$2266 $20,600 21.0 11.0 4.85

talent Search2 academic support program targeted at low income students

$3502 $29,900 73.1 11.7 3.34

Small Schools of Choice3

Small academically non-selective public high-school

$6180 $71,860 59.3 8.6 1.39

Table 3. additional high-school graduates from three interventions.

Note: Calculations based on Column (e) in table 2.

$11.5 billion investment average 204,161 Career academies 558,252 talent search 384,209 Small schools of Choice 160,032

CRITICAL PUBLIC HEALTH 357

assistance (e.g. Temporary Assistance for Needy Families [TANF]; Special Supplemental Nutrition Program for Women, Infants, and Children [WIC]; housing vouchers). Lifetime welfare costs for dropouts are, on average, $9643 more than for graduates (Belfield & Levin, 2007a, p. 60). Because dropouts are more likely to commit crimes, lifetime criminal justice system costs for dropouts exceed those of graduates by $40,701 on average (Belfield & Levin, 2007a, p. 58). Better health status and better employment prospects among high-school graduates results in a 50% decrease in Medicaid enrollment and an average of $74,969 less lifetime spending on Medicare and Medicaid compared to a dropout (Belfield & Levin, 2007a, pp. 54–55). Set against these government cost savings is additional government spending on high-school education, estimated at $39,083 (Belfield & Levin, 2007a, p. 49). (Note that amount is distinct from the cost of the graduation-promotion intervention and represents only the resource cost of the additional education that the intervention is meant to encourage).

As shown in Table 4, the average government savings net of education costs for the entire cohort would amount to approximately $17 billion per year. Combined with increased tax revenue, the total net benefit of an investment in the intervention of average effectiveness would be over $30 billion. All interventions studies have positive benefit–cost ratios, indeed extremely favorably so.

Health and well-being benefits of education interventions

In addition to the financial returns to government, the health benefits of an $11.5 billion investment in dropout prevention interventions are considerable. A study that examined health benefits of high- school graduation found that, compared to a high-school dropout, a high-school graduate could expect a mean improvement of 2.2 quality-adjusted life years (QALYs) by age 65 and 5.1 QALYs by age 85 (Groot & van den Brink, 2004; Muennig & Woolf, 2007). These improvements come through a variety of channels, including better capacity to afford needed medicine and healthy diets, better access to fitness options, and more options for buffering the effects of ill health when it occurs. For the national cohort of students potentially affected by the average intervention, this would result in a gain of nearly 500,000 QALYs by age 65 and just over 1.0 million QALYs by age 85.

Implications

Rapidly rising pharmaceutical prices are not the only source of zero-marginal-value medical care, but they are one expenditure category under policy control that could, if redirected, result in substantial increases in health and quality of life.

In the past, some researchers have suggested that limiting drug price increases to general inflation rates from the 1950s to the 1990s would have resulted in a 30% decrease in research and development (R&D) spending and 330 to 365 fewer new drugs between 1980 and 2001 (Giaccotto, Santerre, & Vernon,

Table 4. Government fiscal returns from an investment of $11.5 billion to promote high-school graduation (in millions of 2016 dollars).

Notes: Columns (a) and (b) are calculated as the numbers of increased graduates reported in table 3 times the estimates provided in the text. total benefit in Column (C) is the sum of Columns (a) + (b). Net benefit in Column (e) is Column (C) − Column (D). internal benefit-Cost ratio in Column (F) is (C)/(D).

(A) (B) (C) (D) (E) (F)

Increased tax revenue

Savings from averted social service costs

net of education costs Total benefit

Program costs Net benefit

Internal benefit cost

ratio average $26,315 $17,605 $43,920 $11,500 $32,420 3.82 Career acade-

mies $71,955 $48,138 $120,093 $11,500 $108,593 10.44

talent search $49,522 $33,130 $82,652 $11,500 $71,152 7.19 Small schools of

choice $20,627 $13,800 $34,427 $11,500 $22,927 2.99

J. E. FIELDING ET AL.358

2005). On the whole, however, empirical evidence on the effect on R&D expenditures of price reductions at the margin is not conclusive (Reinhardt, 2001). For example, one recent study finds a strong effect of Medicare Part D spending on new development, but also notes that the effect was limited to certain classes of drugs (Blume-Kohout & Sood, 2013).

As that study indicates, the real-world processes around drug research funding are complex, and involve balancing multiple objectives, even within pharmaceutical companies. All the same, even sup- posing that modest price reductions did suppress pharmaceutical innovation at the margin, economic theory suggests that the drugs not developed would be those of the least marginal therapeutic benefit. Because, as critics have argued, pharmaceutical R&D expenditures disproportionately go toward me-too drugs with little or no incremental therapeutic benefit (Light, 2009; Light & Lexchin, 2012; QuintilesIMS, n.d.), this marginal therapeutic loss would likely be extremely small. By contrast, the health benefits of alternative uses of this money are large.

Part of the value of this research is to set a target for expected returns on investment. Rather than a simple assertion by the pharmaceutical industry that reducing its profits would suppress innovation it should demonstrate that it can provide greater health benefits than the alternative uses, including our estimates for the educational intervention we have modeled in this paper.

Notwithstanding profit margins that average 18% in the pharmaceutical sector (Anderson, 2014; Reinhardt, 2001) – or indeed because of those high profits – changing policy to reign in pharmaceu- tical prices would be a heavy lift politically, and would encounter exceptionally stiff opposition from both pharmaceutical companies and their well-financed lobbyists. At the same time, there is both substantial public concern about the high price of many drugs (Alpern, Stauffer, & Kesselheim, 2014; Bach, 2015; The Editorial Board, 2015; Pollack, 2016) and a recognition among many policy-makers that with monopoly power comes responsibility to further the public good. A discussion of specific policy options for pharmaceuticals is outside the scope of this paper, but policy proposals exist and has been extensively discussed in the academic literature (Fellows & Hollis, 2013; Stabile et al., 2013). This analysis contributes to what should be a well-informed public discussion of pharmaceutical prices by highlighting the very real opportunity cost of inaction.

Finally, while it is of course not possible to redistribute funds directly from CMS to the Department of Education, that is also not the point. It is inherent in the nature of opportunity costs that they are hypothetical. An opportunity cost is simply one way of expressing the cost of not doing something. The 108th Congress chose to explicitly prohibit any negotiation with pharmaceutical manufacturers over price – an extremely expensive policy choice. Negotiating pharmaceutical prices would be painful to their manufacturers. But cutting prices by the amount we suggest in this example would represent only a 6.5% price reduction and limited to 80 drugs. This implies an annual subsidy of $11.5 billion to the pharmaceutical industry. An alternative use of an $11.5 subsidy would cover 40% of high-school students for one year with dropout prevention programs that would allow 200,000 children to graduate who otherwise would not, and would create nearly 1 million QALYs. Even if the trade-off between these alternatives is indirect and abstract, it is meaningful.

At the state level, however, the trade-off is often real and immediate. Rising pharmaceutical prices have been identified by the Executive Director of the National Association of Medicaid Directors as ‘one of the key factors straining state Medicaid budgets’ (Barlas, 2015). And because states must balance their annual budgets, when spending on Medicaid increases, spending on other social services generally decreases (Tran et al., 2017), with education particularly hard hit (Fossett & Burke, 2004).

In this analysis, a modest change to prices of drugs could produce revenue that could yield an additional 200,000 high-school graduations per year. These numbers represent only a small fraction of the 1.2 million American high-school students who drop out of school annually. Education programs to increase high-school graduation rates are not a panacea. For the average program described above, the number needed to treat is 7.8, and the $11.5 billion per year would be enough to expose only 1.6 million students per year to the intervention, which is about 40% of the students in public schools at each year of high-school. Only one in six students at risk would graduate who otherwise would not

CRITICAL PUBLIC HEALTH 359

have. However, an increase in high-school graduates even of this magnitude could generate additional government savings of tens of billions of dollars, while also substantially improving health outcomes.

An important part of the value of this analysis is to highlight that in education (as in many other social sectors) there are very high expected returns for incremental increases in public spending. The debate should not be whether reducing pharmaceutical prices at the margin reduces potential therapeutic advances, but whether these hypothetical therapeutic advances have greater health and economic returns than other investments, such as providing effective counseling for high-school completion.

Of course, estimates of future cost savings are subject to enormous uncertainty as the policy envi- ronment and underlying risk factors change, and because these uncertainties are themselves uncertain, there is no way to calculate reasonable confidence intervals. There are also geographic variations around these estimates, which are calculated for the State of California. Estimates of local and state savings would be different for different states. Nonetheless, the magnitude of the effects is important, even if too high or too low by a moderate amount.

The method used here of identifying zero-marginal-value pharmaceutical spending is a conservative one. Other methods have produced estimates as much as 4 times as high as these (Yu et al., 2017) or even higher (Light, 2017). Higher estimates of the zero-marginal-value spending on pharmaceuticals, would imply that much more money could be freed up for spending on other policy priorities (Bradley, Elkins, Herrin, & Elbel, 2011), and would of course imply a commensurately higher estimate of the total health opportunity cost of high pharmaceutical spending.

As long as US policy-makers continue to make investments in low-value activities while neglecting high-value ones, the US health system will continue to underperform its peers.

Note 1. The case of Sovaldi, a drug that cures Hepatitis C, helps clarify what drives and does not drive increased costs

due to price increases. Sovaldi was introduced to the market with a very high price per unit, for which it garnered considerable press coverage. But Sovaldi also has a very high therapeutic value, even relative to price. Moreover, its price was virtually unchanged between 2013 and 2014. Total drug spending increased by the highest dollar amount between 2013 and 2014, in part because of rapidly increasing unit sales of Sovaldi after its introduction in late 2013. Yet the simulated increment due to price increases was not the highest between 2013 and 2014 and had nothing to do with Sovaldi. Although Sovaldi has garnered considerable attention for its high price, the analysis here takes no position on whether the price of Sovaldi when it was introduced may in fact be economically justifiable. Instead, the focus is on price increases after drugs are already introduced.

Disclosure statement No potential conflict of interest was reported by the authors.

References Aitken, M., Berndt, E. R., Cutler, D., Kleinrock, M., & Maini, L. (2016). Has the era of slow growth for prescription drug spending

ended? Health Affairs, 35(9), 1595–1603. Alpern, J. D., Stauffer, W. M., & Kesselheim, A. S. (2014). High-cost generic drugs – implications for patients and policymakers.

New England Journal of Medicine, 371(20), 1859–1862. Anderson, R. (2014, November). Pharmaceutical industry gets high on fat profits. BBC Online. Retrieved from http://www.

bbc.com/news/ Bach, P. B. (2015, January). Why drugs cost so much. The New York Times. Retreived from https://www.nytimes.com/ Baker, D. (2006). The savings from an efficient Medicare prescription drug plan. Center for Economic and Policy Research.

Retrieved from http://cepr.net/documents/efficient_medicare_2006_01.pdf Barlas, S. (2015). States try to control Medicaid pharmaceutical costs: Numerous, diverse cost pressures force myriad reform

efforts. Pharmacy and Therapeutics, 40(4), 260. Belfield, C. R., & Levin, H. M. (2007a). The economic losses from high school dropouts in California. Santa Barbara, CA: California

Dropout Research Project, University of California. Retrieved from http://cdrpsb.org/download.php?file=policybrief1.pdf

J. E. FIELDING ET AL.360

http://www.bbc.com/news/
http://www.bbc.com/news/
https://www.nytimes.com/
http://cepr.net/documents/efficient_medicare_2006_01.pdf
http://cdrpsb.org/download.php?file=policybrief1.pdf
Belfield, C. R., & Levin, H. M. (2007b). The return on investment for improving California’s high school graduation rate. Santa Barbara, CA: California Dropout Research Project, University of California. Retrieved from http://cdrpsb.org/download. php?file=researchreport2.pdf

Bloom, H. S., & Unterman, R. (2012). Sustained positive effects on graduation rates produced by New York City’s small public high schools of choice. New York, NY: MDRC. Retrieved from https://www.mdrc.org

Blume-Kohout, M. E., & Sood, N. (2013). Market size and innovation: Effects of Medicare Part D on pharmaceutical research and development. Journal of Public Economics, 97, 327–336.

Bradley, E. H., Elkins, B. R., Herrin, J., & Elbel, B. (2011). Health and social services expenditures: Associations with health outcomes. BMJ Quality & Safety, 20(10), 826–831.

Brunello, G., Fort, M., Schneeweis, N., & Winter-Ebmer, R. (2016). The causal effect of education on health: What is the role of health behaviors?”. Health Economics, 25(3), 314–336.

Bureau of Labor Statistics. (2016). Consumer price index – All urban consumers – Medical Care (Series CUUR0000SAM). Retrieved June 17, 2016, from http://data.bls.gov/timeseries/CUUR0000SAM?output_view=pct_12mths

Centers for Medicare and Medicaid Services. (2015a). Medicare drug spending dashboard. Retrieved June 17, 2016, from https://www.cms.gov/Newsroom/MediaReleaseDatabase/Fact-sheets/2015-Fact-sheets-items/2015-12-21.html

Centers for Medicare and Medicaid Services. (2015b). National health expenditure accounts. Retrieved February 24, 2017, from https://www.cms.gov/research-statistics-data-and-systems/statistics-trends-and-reports/nationalhealthexpenddata/ downloads/highlights.pdf

Conti, G., & Heckman, J. J. (2013). The developmental approach to child and adult health. Pediatrics, 131(Suppl. 2), S133–S141. Conti, G., Heckman, J., & Urzua, S. (2010). The education-health gradient. The American Economic Review, 100(2), 234–238. The Editorial Board.(2015). No justification for high drug prices. The New York Times. Retrieved from https://www.nytimes.

com/ Fellows, G. K., & Hollis, A. (2013). Funding innovation for treatment for rare diseases: Adopting a cost-based yardstick

approach. Orphanet Journal of Rare Diseases, 8(1), 180. Fossett, J. W., & Burke, C. (2004). Medicaid and state budgets in FY 2004: Why Medicaid is so hard to cut. Albany, NY: The Nelson

A. Rockefeller Institute of Government. Giaccotto, C., Santerre, R. E., & Vernon, J. A. (2005). Drug prices and research and development investment behavior in the

pharmaceutical industry. The Journal of Law & Economics, 48(1), 195–214. Groot, W., & van den Brink, H. M. (2004). The health effects of education: Survey and meta-analysis. Retrieved from https://

www.researchgate.net/ Gubits, D., Shinn, M., Bell, S., Wood, M., Dastrup, S., Solari, C. D., … Abt Associates, Inc. (2015). Family options study: Short-

term impacts of housing and services interventions for homeless families. U.S. Department of Housing and Urban Development, Office of Policy Development and Research. Retrieved from https://www.huduser.gov/portal/sites/ default/files/pdf/FamilyOptionsStudy_final.pdf

The Guide to Community Preventive Services. (2013). Promoting health equity through education programs and policies: High school completion programs. Retrieved June 20, 2016, from http://www.thecommunityguide.org/healthequity/ education/highschoolcompletion.html

Haines, A., Kovats, R. S., Campbell-Lendrum, D., & Corvalan, C. (2006). Climate change and human health: Impacts, vulnerability and public health. Public Health, 120(7), 585–596.

Hutton, D., Newman-Casey, P. A., Tavag, M., Zacks, D., & Stein, J. (2014). Switching to less expensive blindness drug could save Medicare Part B $18 billion over a ten-year period. Health Affairs, 33(6), 931–939.

Kantarjian, H., & Zwelling, L. (2013). Cancer drug prices and the free-market forces. Cancer, 119(22), 3903–3905. Kemple, J. J. (2008). Career academies: Long-term impacts on labor market outcomes, educational attainment, and transitions

to adulthood. New York, NY: MDRC. Retrieved from https://www.mdrc.org Levin, H. M., Belfield, C., Hollands, F., Bowden, A. B., Cheng, H., Shand, R., … Hanisch-Cerda, B. (2012). Cost-effectiveness

analysis of interventions that improve high school completion. New York: Center for Benefit-Cost Studies of Education, Teachers College, Columbia University.

Light, D. W. (2009). Global drug discovery: Europe is ahead. Health Affairs, 28(5), w969–w977. Light, D. W. (2017). Debunking the pharmaceutical research ‘free rider’ myth: A response to Yu, Helms, and Bach. Health

Affairs Blog. Retrieved from https://www.healthaffairs.org/ Light, D. W., & Kantarjian, H. (2013). Market spiral pricing of cancer drugs. Cancer, 119(22), 3900–3902. Light, D. W., & Lexchin, J. (2012). Pharmaceutical research and development: What do we get for all that money? BMJ, 345,

e4348. Maxwell, N. L., & Rubin, V. (2000). High school career academies: A pathway to educational reform in urban school districts?.

W. E. Upjohn Institute. Retrieved from http://research.upjohn.org/ Muennig, P., & Woolf, S. H. (2007). Health and economic benefits of reducing the number of students per classroom in US

primary schools”. American Journal of Public Health, 97(11), 2020–2027. The Office of the Actuary in the Centers for Medicare & Medicaid Services. (2017). National health expenditure historical and

projected. Baltimore, MD: U.S. Department of Health and Human Services. Pollack, A. (2016, July). Makers of Humira and Enbrel using new drug patents to delay generic versions. The New York Times.

Retrieved from https://www.nytimes.com/

CRITICAL PUBLIC HEALTH 361

http://cdrpsb.org/download.php?file=researchreport2.pdf
http://cdrpsb.org/download.php?file=researchreport2.pdf
https://www.mdrc.org
http://data.bls.gov/timeseries/CUUR0000SAM?output_view=pct_12mths
https://www.cms.gov/Newsroom/MediaReleaseDatabase/Fact-sheets/2015-Fact-sheets-items/2015-12-21.html
https://www.cms.gov/research-statistics-data-and-systems/statistics-trends-and-reports/nationalhealthexpenddata/downloads/highlights.pdf
https://www.cms.gov/research-statistics-data-and-systems/statistics-trends-and-reports/nationalhealthexpenddata/downloads/highlights.pdf
https://www.nytimes.com/
https://www.nytimes.com/
https://www.researchgate.net/
https://www.researchgate.net/
https://www.huduser.gov/portal/sites/default/files/pdf/FamilyOptionsStudy_final.pdf
https://www.huduser.gov/portal/sites/default/files/pdf/FamilyOptionsStudy_final.pdf
http://www.thecommunityguide.org/healthequity/education/highschoolcompletion.html
http://www.thecommunityguide.org/healthequity/education/highschoolcompletion.html
https://www.mdrc.org
https://www.healthaffairs.org/
http://research.upjohn.org/
https://www.nytimes.com/
QuintilesIMS. (n.d.) Total nominal spending on medicines in the U.S. from 2002 to 2016 (in billion U.S. dollars). In Statista – The Statistics Portal. Retrieved June 17, 2016, from https://www.statista.com/statistics/238689/us-total-expenditure- on-medicine/

Reinhardt, U. E. (2001). Perspectives on the pharmaceutical industry. Health Affairs, 20(5), 136–149. Ross, C. E., & Wu, C. (1995). The links between education and health. American Sociological Review, 60(5), 719–745. Ruhm, C. J. (2000). Parental leave and child health. Journal of Health Economics, 19(6), 931–960. Scobie, J. (2015). Global agewatch index 2015. Age International, HelpAge Global Network. Retrieved from https://www.

ageinternational.org.uk/ Shih, C., Schwartz, J., & Coukell, A. (2016). How would government negotiation of Medicare Part D drug prices work? Health

Affairs Blog. Retrieved June 17, 2016, from http://healthaffairs.org/blog/ Stabile, M., Thomson, S., Allin, S., Boyle, S., Busse, R., Chevreul, K., … Mossialos, E. (2013). Health care cost containment

strategies used in four other high-income countries hold lessons for the United States. Health Affairs, 32(4), 643–652. Tanaka, S. (2005). Parental leave and child health across OECD countries. The Economic Journal, 115(501), F7–F28. Tanenbaum, S. J. (2017). Can payment reform be social reform? The lure and liabilities of the ‘Triple Aim’. Journal of Health

Politics, Policy and Law, 42(1), 53–71. Thorpe, K. E. (2005). The rise in health care spending and what to do about it. Health Affairs, 24(6), 1436–1445. Tran, L. D., Zimmerman, F. J., & Fielding, J. E. (2017). Public health and the economy could be served by reallocating medical

expenditures to social programs. SSM – Population Health, 3, 185–191. Wilson, S. J., Tanner-Smith, E. E., Lipsey, M. W., Steinka-Fry, K., & Morrison, J. (2011). Dropout prevention and intervention

programs: Effects on school completion and dropout among school-aged children and youth. Retrieved April 19, 2016, from http://campbellcollaboration.org/lib/project/158/

Young, P. L., & Olsen, L. (2010). The healthcare imperative: Lowering costs and improving outcomes. Washington, DC: The National Academies Press, Institute of Medicine.

Yu, N., Helms, Z., & Bach, P. (2017). R&D costs for pharmaceutical companies do not explain elevated US drug prices. Health Affairs Blog. Retrieved from https://www.healthaffairs.org/

J. E. FIELDING ET AL.362

https://www.statista.com/statistics/238689/us-total-expenditure-on-medicine/
https://www.statista.com/statistics/238689/us-total-expenditure-on-medicine/
https://www.ageinternational.org.uk/
https://www.ageinternational.org.uk/
http://healthaffairs.org/blog/
http://campbellcollaboration.org/lib/project/158/
https://www.healthaffairs.org/
Copyright of Critical Public Health is the property of Routledge and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.

Abstract
Introduction
Increases in Medicare drug prices
Education as an alternative investment
Financial returns to government of educational investments
Health and well-being benefits of education interventions
Implications
Note